Software to compute sample size for high-volume genetic studies

DESCRIPTION (provided by applicant): In the study of complex traits such as cancer, heart disease, and asthma, it is essential for investigators to have an honest estimate of required sample size when they design their studies. This is particularly true in the modern era of high-volume genetic studies, including pathway-based candidate-gene and genome-wide association studies, as they typically require large sample sizes and therefore can be quite expensive to conduct. The overall aim of our Phase I and Phase II projects will be to develop a comprehensive software program (QuantoHV) for the design of High Volume association studies of genes, environmental factors, gene- environment (GWE) interaction, and gene-gene (GWG) interaction. In developing QuantoHV, we will account for issues that arise in the design of high-volume studies of complex traits, including multiple comparisons, one- versus two-stage designs, population stratification, choice of study design (e.g. case-control, cohort, family-based), and the use of existing genetic data resources (e.g. HapMap). In this Phase I application, we will achieve the following aims: 1. Determine which specific features should be included in QuantoHV, taking into account current and future trends in genetic research. 2. Determine whether to develop QuantoHV as a Windows or a Web-based application. 3. Investigate alternative models for commercialization and distribution of QuantoHV The investigators on this project have a long history of working together to develop user-friendly software for genetic-epidemiology studies. This, and their collective experience in designing and carrying out high-volume genetic studies, makes this study team uniquely qualified to develop a software program that will fill an important void and ultimately will lead to better-designed studies of complex traits. PUBLIC HEALTH RELEVANCE: There are currently no software programs that compute sample size or power for the types of studies currently being proposed by many investigators, specifically those that utilize high-volume genetic data such as candidate-gene and genome-wide association studies. Successful completion of our Phase I aims will put us in position to develop a unique and novel software program that will fill this void.